Shortcode Tokenizer and AST

Tokenizes a string containing shortcodes (re-popularized by WordPress) and outputs
it as an AST that can be used for further parsing.
If you are only looking for simple transformations from a shortcode to a string
I suggest that you give these libs a try:
If you need more control this is the lib for you.
Install
npm install shortcode-tokenizer
Usage
import ShortcodeTokenizer from 'shortcode-tokenizer'
const input = `
<h1>Cool Shop</h1>
[row]
[col width=6 class="featured"]
[product-list list="featured" /]
[/col]
[col width=6 class="featured"]
<div>Ad: Buy more, Buy often!</div>
[/col]
[/row]
`
const tokenizer = new ShortcodeTokenizer(input)
tokenizer.ast()
The AST outputted is at root level an array of two nodes, a text node and a
code-node:
[
{
type: 'TEXT',
body: "<h1>Cool Shop</h1>"
pos: 0
},
{
type: 'OPEN',
name: 'row',
pos: 19,
body: '[row]',
isClosed: true,
params: {},
children: [
... a whitespace TEXT token ...,
{
type: 'OPEN',
name: 'col',
pos: 27,
body: '[col width=6 class="featured"]',
isClosed: true,
params: {
width: 6,
class: 'featured'
},
children: [ ... and so on ... ]
}
]
}
]
(Note: the same data-structure is used to represent tokens from the lexing and
nodes in the AST, see CLOSE, below)
There are 5 token types:
- TEXT: plain text.
body
contains the content.
- OPEN: an open token, e.g.
[row]
.
- SELF_CLOSING: a self-closing token
[post id=1/]
.
- CLOSE: a close token, e.g.
[/row]
. You will only see the left-over of these
tokens in the AST as OPEN tokens that have their isClosed
value set to
true.
- ERROR: when in non-strict mode offending tokens are converted to ERROR tokens
and which behaves like TEXT nodes.
API
Property options.strict
By default strict-mode is enabled and syntax errors will be thrown. Setting
strict-mode to false will convert all errors into ERROR nodes.
Property options.skipWhiteSpace
By default skipWhiteSpace is off. When turned on all whitespace is trimmed and
if there is nothing left the TEXT node is skippid.
constructor and input()
You can pass input in the constructor you can set it later using input()
.
Examples:
let t = new ShortcodeTokenizer('[code][/code]')
t.ast()
t.input('[code][/code]')
t.ast()
t.ast('[code][/code]')
tokens()
Returns all tokens.
ast([input])
Returns an AST created from the input.
buildTemplate([Token token], [object|string|null params=null], [bool deep=false])
Builds template on given token. You can overwrite params in root param by params
and also in all nested setting deep="true"
param.
Changelog
See CHANGELOG.md